PhD Students (internal and external, supervision and collaboration)
Reut Tsarfaty (UvA, NWO MOSAIEK project): graduation around March/April 2009
2007-2010 Supervision Markos Mylonakis (UvA, NWO VIDI financed project)
Hany Hassan (Dublin City University) PROSPECT project (together with prof. Andy Way): graduation Nov/Dec 2008
Ventsislav Zhechev (Dublin City University) ATTEMPT project (together with prof. Andy Way and dr. Mary Hearne)
Sep/Oct 2008 - 2010: Tejaswini Deoskar (0.5fte postdoc and 05fte lecturer substituting for myself in teaching Stat. NLP courses).
Vacancy (2 years -- Project Priors for Estimating Probabilistic Grammars from Incomplete Data )
Temporarily Yoav Seginer -- substituting lecturer for my statistical NLP courses
Detlef Prescher (2003-2006 -- Project LeStoGram Consistent Statistical Estimation from Treebanks)
General Description of Research Interest
Priors for Estimating Probabilistic Grammars from Incomplete Data
(NWO-EW - VIDI) [2007-2011]
0.5 FTE lecturer 4.5 years (Yoav Seginer)
1.0 FTE postdoc position 3 years (vacancy)
1.0 FTE PhD position (Markos Mylonakis)
All-fragments Approaches and Statistical Machine Translation
'ATTEMPT: "All Trees" Efficient Models of Parsing and Translation'.
In collaboration with DCU people: Andy Way, Mary Hearne, Hany Hassan, John Tinsley and Ventsislav Zhechev.
Statistical Estimation from Treebanks (CLOSED)
NWO-EW Open Competitie [2003-2006]
Learning Stochastic Tree-Grammars from Treebanks (LeStoGram)
People involved: Detlef Prescher (postdoc), Remko Scha and Khalil Sima'an (leaders).
Graduates: Andreas Zollmann, Linh Thuy Nguyen and Luciano Buratto.
PROJECT PERIOD FINISHED WITH FINAL REPORT
Modern Hebrew and Arabic Processing [2000-present]
Joint Modeling and Estimation of Probabilistic Models for Hebrew Morphological and Syntactic Processing
People involved: Khalil Sima'an (UvA), Reut Tsarfaty (UvA), and Hebrew Project (Technion)
Roy-Bar Haim (Technion, Haifa), Saib Mansour (Technion, Haifa)
Collaboration: 2005 Johns Hopkins University (JHU) Summer Workshop 2005 on Parsing Arabic Dialect.
Beyond Treebank Annotation: Ambiguity Resolution by Similarity-Based Performance Models
Personal innovation grant (KNAW Fellowship, Royal Dutch Academy for Sciences and Humanities), 2002.
Funding of Project ended in 2003 when I accepted Assistant Professorship at UvA.
Data Oriented Parsing
An ongoing joint project with various colleagues, specifically Remko Scha, Rens Bod, Andy Way, Mary Hearne and many others.
ONGOING: See Remko Scha and his former student Rens Bod for more on this
- Statistical Machine Translation
- Structured Statistical Memory-Based Models
- Statistical Parsing
- Probabilistic models and their applications
- Statistical Estimators
- Machine Learning techniques
- Applied Natural Language Processing
- Language and Speech Technology
Computational models of intelligent behavior, especially language processing
- Stochastic ambiguity resolution using probabilistic models of Natural Language Processing
- Machine Learning: Estimation methods
- Applications e.g. Speech Understanding, Question Answering, Machine Translation
- Algorithms for and computational complexity of stochastic parsing
- Applications of machine learning algorithms to cognitive modeling
- Statistical and other Machine Translation Models